During the 1990s, the typical age of diagnosis for autism fell, from almost 6 years to 3 years of age. Although early diagnosis is widely viewed as essential to the most effective interventions and treatments, there is substantial variation in who is diagnosed early or later. In this paper, we explore the characteristics of children, families, and communities that are associated with diagnostic timing, and how these have changed between the 1992 and 2001 birth cohorts. We find that the most robust determinant of early diagnosis is parental education, and that although the gap between high and low SES children has diminished over the decade, it remains significant.
In 1994, Californians passed Proposition 187, a ballot initiative aimed at keeping undocumented immigrants and their children from using publicly funded social services. We link the passage and overturning of Prop. 187 to observed patterns of enrollment of the children of Latina immigrants in the Department of Developmental Services (DDS) autism caseload over time. Although the group of children we examine were US citizens, born in California and fully eligible for DDS services, many of them live in households of mixed immigration status, and we find evidence of a 'chilling effect' on their service utilization. Specifically, after Prop. 187 is passed enrollment is depressed, relative to native-born whites, then increases rapidly after it is overturned in court, only to decline again with heightened post-9/11 immigration enforcement. This suggests that some citizen children with autism living in mixed-status families do not receive, or receive late, the developmental services to which they are entitled. These findings remind us that autism rates are in part a product of the social and political environment, and that immigration policy can have unintended consequences for our most vulnerable citizens.